jax-ml/jax · error · TypeError

expected sequence object with len >= 0 or a single integer

Error message

expected sequence object with len >= 0 or a single integer

What it means

jnp.zeros rejects generator objects for shape. Generators are single-use and have no reliable length, so they cannot describe a static array shape under JAX's shape canonicalization; NumPy's more permissive behavior is deliberately not mirrored.

Source

Thrown at jax/_src/numpy/array_creation.py:88

    Array of the specified shape and dtype, with the given device/sharding if specified.

  See also:
    - :func:`jax.numpy.zeros_like`
    - :func:`jax.numpy.empty`
    - :func:`jax.numpy.ones`
    - :func:`jax.numpy.full`

  Examples:
    >>> jnp.zeros(4)
    Array([0., 0., 0., 0.], dtype=float32)
    >>> jnp.zeros((2, 3), dtype=bool)
    Array([[False, False, False],
           [False, False, False]], dtype=bool)

  .. _explicit sharding: https://docs.jax.dev/en/latest/parallel.html
  """
  if isinstance(shape, types.GeneratorType):
    raise TypeError("expected sequence object with len >= 0 or a single integer")
  if (m := _check_forgot_shape_tuple("zeros", shape, dtype)): raise TypeError(m)
  dtype = dtypes.check_and_canonicalize_user_dtype(
      float if dtype is None else dtype, "zeros")
  shape = canonicalize_shape(shape)
  sharding = util.choose_device_or_out_sharding(
      device, out_sharding, 'jnp.zeros')
  return lax.full(shape, 0, dtype, sharding=sharding)


@export
def ones(shape: Any, dtype: DTypeLike | None = None, *,
         device: xc.Device | Sharding | None = None,
         out_sharding: NamedSharding | P | None = None) -> Array:
  """Create an array full of ones.

  JAX implementation of :func:`numpy.ones`.

  Args:

View on GitHub (pinned to 1e1c6a8fc0)

Solutions

1. Materialize the generator: jnp.zeros(tuple(gen)) or jnp.zeros([*gen])
2. Prefer explicit tuples: jnp.zeros((2, 3))

Example fix

# before
a = jnp.zeros(d for d in [2, 3])
# after
a = jnp.zeros(tuple(d for d in [2, 3]))  # or simply (2, 3)

import types
def canonical(shape):
    return tuple(shape) if isinstance(shape, types.GeneratorType) else shape

null

null

When it happens

Trigger: jnp.zeros((n) for n in dims) or jnp.zeros(range(3)) — wait, range passes; specifically passing a types.GeneratorType like jnp.zeros(x for x in [2,3]).

Common situations: Refactoring NumPy code that computed shapes lazily; passing a generator expression where a tuple was intended, often due to a trailing comma typo or comprehension misuse.

Related errors

• scan got `length` argument of {} which disagrees with leadin
• conv_general_dilated batch_group_count must divide lhs batch
• conv_general_dilated rhs output feature dimension size must
• conv_general_dilated window and window_strides must have the
• Wrong number of explicit pads for convolution: expected {},

AI-assisted analysis of jax-ml/jax@1e1c6a8fc0 (2026-08-27). Data as JSON: /api/errors/6f10adc922c4e450. Report an issue: GitHub.